LAUSR

AbstractPolicymakers have been considering vehicle and fuel taxes to reduce transportation greenhouse gas emissions, but there is little evidence on the relative efficacy of these approaches. We examine an annual vehicle registration tax, the vehicle excise duty (VED), which is based on carbon emissions rates. The UK first adopted the system in 2001 and made substantial changes to it in the following years. Using a highly disaggregated dataset at the trim-variant level of UK registrations and characteristics of new cars, we estimate the effect of the VED on new vehicle registrations and their carbon emissions. The VED increased the adoption of low-emissions vehicles and discouraged the purchase of very polluting vehicles, but it had a small effect on aggregate emissions. Using the empirical estimates, we compare the VED with two hypothetical taxes: a tax proportional to carbon emissions per kilometer, and a carbon tax. The VED reduces total emissions from new cars twice as much as the emissions rate tax but by half as much as the emissions tax. Much of the advantage of the emissions tax arises from adjustments in miles driven, rather than the composition of the new car sales.